[TriEmbed] efficiency of power supply + thingy
Jim Sugaroo
jim.sugaroo at gmail.com
Tue Jun 24 20:41:49 CDT 2014
Chip!
I'm used to seeing power supply efficiency as a function of load and mainly
when load is unknown such as a generic power supply. I don't think you
really need such a measurement over various load currents for your thingy.
I think your thingy has a specific load profile and you want to use it for
your test. I'm thinking you want to tweak your circuits, evaluate other
power supplies, change software, etc to get more *relative *battery life.
Here's my suggestion to evaluate changes and how efficient your power
supply + thingy operate:
1. Use the actual load of the micro and circuitry but have it wake up more
often and take a measurement, or not go to sleep, making constant
measurements. The goal here is to increase load for shorter test times but
keep same load profile. Depending on #4, you may not need to change
anything.
2. Have the thingy output a count value that is the count since power ON
and record this count value.
3. I'm assuming your have an under voltage lockout or some hysteresis in
your power supply such that it doesn't stutter as input voltage gets too
low.
4. Use a BFC as a power source, charge it up, with to Vin with SPST switch
(and resistor), then discharge it into your power supply that feeds the
thingy. Try a 10,000uF or do some time math to target a desired run time.
5. Measure the count value until it stops counting - when the voltage in
BFC is too low. Run it 5X and make sure its consistent - take an average
of the run times for a specific configuration.
You can use this count value to evaluate how efficient changes in your
design are. This will be a faster test than your actual application. Then
you can do long term testing using a real battery and real application but
it should correlate to this fast efficiency test. You could also calculate
energy/run-time which is an efficiency metric. The voltage decay of the
battery will not be the same as for a capacitor but I don't think that
matters much for your test.
Jim Sugaroo
-------------------
---------------
-----------
--- ---
-- --
-- --
- -
On Tue, Jun 24, 2014 at 1:00 PM, <triembed-request at triembed.org> wrote:
> Send TriEmbed mailing list submissions to
> triembed at triembed.org
>
> To subscribe or unsubscribe via the World Wide Web, visit
> http://mail.triembed.org/mailman/listinfo/triembed_triembed.org
> or, via email, send a message with subject or body 'help' to
> triembed-request at triembed.org
>
> You can reach the person managing the list at
> triembed-owner at triembed.org
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of TriEmbed digest..."
>
>
> Today's Topics:
>
> 1. Power Supply Testing - Am I over or under thinking this?
> (Charles McClelland)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Tue, 24 Jun 2014 11:29:13 -0400
> From: Charles McClelland <chip at mcclellands.org>
> To: TriEmbed Discussion <triembed at triembed.org>
> Subject: [TriEmbed] Power Supply Testing - Am I over or under thinking
> this?
> Message-ID: <D55875FF-1168-4B93-8026-22727767AC2E at mcclellands.org>
> Content-Type: text/plain; charset="windows-1252"
>
> To all,
>
> In my never ending pursuit of longer battery life and / or less batteries,
> I have been exploring moving from a linear DC-DC converter to a switching
> power converter.
>
> TI was gracious enough to send me some samples of their TPS63031
> Buck-Boost Switching Power Supply which requires few external components
> and with their WEBench and Switcher Pro tools, Pete and I were able to
> create a simple board to test this chip. Pete lent his expertise and
> lessons learned in working with these tiny QFN packages and we were able to
> build one of these boards over the weekend.
>
> Now I need to test it. I wanted to share my approach with this group to
> make sure I am headed down the right track.
>
> Major components:
> - The TPS63031 which should maintain a constant 3.3V output as the
> batteries drain from 5.5V to 1.8V
> - A data logger
> - Arduino (5V)
> - Battery packs of various types up to 5.5V
> - Some sort of load (like a 13.2 ohm 1W resistor)
> - A ?high-side? current sensor like this one from Adafruit which
> has an i2C output
> - A Real Time Clock like the DS1307
>
> Here is the under / over thinking part. In the end, I want to have a
> rough idea how efficient this device is and whether is can maintain a
> constant 3.3V voltage and 250mA current while the batteries run down.
>
> My proposed approach is:
> - Three power rails (Vin for the Power Supply, Vout for the Power
> Supply and 5V for the Arduino) and a common ground
> - A ?load? resistor connecting the Vout to ground
> - A current / voltage sensor to measure on Vin
> - I would then ask the Arduino to log the following every 30
> seconds or so - Vin, Iin and Vout
> - As long as the supply is delivering the 3.3V output, calculating
> Iout would be the voltage divided by the fixed load reisitance
>
> After the batteries die (Vout drops below 3.3V), I could calculate the
> efficiency as the following:
>
> - Instantaneous Efficiency % = (Vout * Iout) / (Vin * Iin) *100
> - I think I could simply integrate the instantaneous efficiency
> over the duration of the test to get the overall efficiency.
>
> Am I thinking of this correctly?
>
> Thanks, Chip
>
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <
> http://mail.triembed.org/pipermail/triembed_triembed.org/attachments/20140624/379e9912/attachment-0001.html
> >
>
> ------------------------------
>
> Subject: Digest Footer
>
> _______________________________________________
> TriEmbed mailing list
> TriEmbed at triembed.org
> http://mail.triembed.org/mailman/listinfo/triembed_triembed.org
>
>
> ------------------------------
>
> End of TriEmbed Digest, Vol 13, Issue 17
> ****************************************
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.triembed.org/pipermail/triembed_triembed.org/attachments/20140624/e8602082/attachment.htm>
More information about the TriEmbed
mailing list